17 



relative dominance. Importance values were used to determine the FTI 

 number for each species. 



When species could not be positively identified in the field, 

 voucher specimens were collected and later identified. Species nomen- 

 clature was determined using the National List of Scientific Plant Names 

 (U.S. Department of Agriculture 1982). 



Calculating Species FTI Numbers 



Changes in composition of biotic communities along environmental 

 gradients can be addressed with several statistical techniques, the most 

 notable being gradient analysis (Whittaker 1978) . Gradient analysis can 

 take several different forms depending on the objective of the analysis. 

 Inferring environmental values (e.g., hydrologic zones) from vegetative 

 species composition is called a "calibration problem" by Ter Braak and 

 Prentice (1988) and is the appropriate approach for this study. 



One method of calibration is to use weighted averaging (WA) to 

 estimate environmental factors at sites based on species optima. If a 

 species exhibits a unimodal distribution with respect to an environmen- 

 tal variable, its occurrence is concentrated around the peak of this 

 function (Ter Braak and Prentice 1988). Species with similar optima 

 will naturally tend to occur together. Therefore, an intuitive estimate 

 of the environmental factor of a site is the average of the optima for 

 the species present. The FTI numbers represent weighted averages of 

 species occurrance . 



Two additional statistical methods of calibrating an environment 

 (hydrologic zone) with vegetation, recommended by Ter Braak and Prentice 



